KR20220018846A - High-focus and high-sensitivity fluorescence detection scanning system equipped with a machine learning-based liquid sample quantitative analysis algorithm - Google Patents

Abstract

The present invention relates to a highly focused and a highly sensitive fluorescence detection scanning system with a machine learning-based liquid sample quantitative analysis algorithm for small quantitative polymerase chain reaction (aPCR) equipment, which detects and analyzes a multi-fluorescence signal on a genetic sample amplified during analysis at high sensitivity to improve a detection limit and sensitivity to improve analysis precision of molecular diagnosis. The fluorescence detection scanning system comprises: a multi-fluorescence detection optical module having a light emission-light detection scan structure; a two-dimensional driving stage supporting planar movement of the multi-fluorescence detection optical module; and a system control board controlling motions of the multi-fluorescence detection optical module and the two-dimensional driving stage.

Description

HIGH-FOCUS AND HIGH-SENSITIVITY FLUORESCENCE DETECTION SCANNING SYSTEM EQUIPPED WITH A MACHINE LEARNING-BASED LIQUID SAMPLE QUANTITATIVE ANALYSIS ALGORITHM

The present invention is a small qPCR (quantitative polymerase chain reaction) equipment that can detect and analyze multiple fluorescence signals with high sensitivity for a gene sample being amplified during the analysis process to improve the detection limit and sensitivity and improve the analysis precision of molecular diagnosis. It relates to a multi-channel, high-speed, high-sensitivity fluorescence detection scanning system equipped with a running-based quantitative analysis algorithm for liquid samples.

Quantitative polymerase chain reaction (quantitative PCR, qPCR, or Real Time PCR) based on semi-automated or automated liquid sample processing technology is a pre-processing process such as incubation, washing, separation and reaction processing, and an analysis process to quantify the concentration of amplification materials It refers to a molecular diagnostic system composed of

Various attempts are being made to develop point-of-care diagnostic devices necessary for immune and molecular diagnostics, centering on domestic biomedical-related venture companies.

In the field of immunodiagnostics, a rapid diagnostic kit applied with a membrane assay (column, lateral flow assay) has been developed, and a small rapid on-site diagnostic device that detects and analyzes the degree of color and fluorescence expression of gold particles in the diagnostic kit is being developed. .

In developing a fully automatic diagnostic system for invasive liquid samples required in the medical field, in order to increase the accuracy and precision of detection of liquid samples, a detection method using liquid rather than sample detection using the existing membrane method is absolutely necessary. In a liquid sample fully automatic diagnosis system, an efficient and simple pre-processing process and a high-accuracy measurement, detection and analysis process must be properly combined and combined.

The conventional conventional chemiluminescence method has the advantage of accumulating and amplifying the signal that the luminescent product changed according to the enzyme amplification reaction is changed, but has a characteristic that the signal generation amount fluctuates according to the minute fluctuation of the amplification reaction of the enzyme. Therefore, there is a disadvantage in that the reliability of the test itself is lost unless a standard curve for quantification is generated by simultaneously detecting standard samples of different concentrations in each detection process.

On the other hand, the fluorescence detection method has a relatively low detection limit compared to the enzyme-based chemiluminescence method, but quantitatively matches the amount of the collected fluorescent label and the fluorescence signal, and there is almost no variation in the amount of fluorescence depending on the surrounding environment. It has the advantage of being able to immediately quantify the value.

On the other hand, it is necessary to develop and mount a fluorescence detection system with advanced performance in diagnostic PCR, such as qPCR, Real-Time PCR, and Digital PCR, to ensure on-site performance and speed.

In addition, since the multiple fluorescence detection system for liquid samples is a core technology element of each diagnostic system developer, existing diagnostic system developers do not openly supply them to the development of other companies' equipment in a universal way, even if they have successful results. Therefore, it is necessary to develop a fluorescence detection system specialized for the measurement detection analysis process to be used by each diagnostic system developer.

As described above, there is a significant need for the development of a fluorescence detection scanning system that can effectively pre-process diagnostic PCR in the current medical field and can be specialized and applied to the measurement detection analysis process of each diagnostic system developer.

Registered Patent Publication No. 10-1847411 (2018.04.04)

The present invention was derived to meet the needs of the prior art, and an object of the present invention is to detect and analyze multiple fluorescence signals with high sensitivity for a gene sample being amplified during the analysis process to improve the detection limit and sensitivity, and to improve molecular diagnosis. The goal is to provide a multi-channel, high-speed, high-sensitivity fluorescence detection scanning system equipped with an algorithm for quantitative analysis of liquid samples based on machine learning for small qPCR (quantitative polymerase chain reaction) equipment that can improve the analysis precision of

A high-speed, high-sensitivity, multi-channel fluorescence detection scanning system according to an aspect of the present invention for solving the above technical problem uses a high-speed, high-light, high-sensitivity optical module based on fine beam spot control. In order to detect emission light from a fluorescent label with high sensitivity, a lens with a high aperture ratio is used to focus the exciting light energy into a fine beam spot and to increase the light collection efficiency. In addition, the optical module system has an optical design structure using a light emitting diode (LED) as a radiation light source, a filter for each wavelength band, a miniature lens, and a high-sensitivity optical sensor to match the optical characteristics of multiple fluorescent label materials, and is applied It is equipped with an optical system and a drive system designed to be flexibly adjusted according to the specifications of liquid-based small and automated equipment.

In one embodiment, the high-speed, high-sensitivity fluorescence detection scanning system uses micro-region segmentation scanning and detection density increase technology. This technology detects fluorescence signals by scanning the measurement target area in a micro-area division method, which is subdivided into minute units, rather than detecting the fluorescent material in a certain volume of a liquid sample with a single measurement.

In one embodiment, the high-speed, high-sensitivity fluorescence detection scanning system uses machine learning-based quantitative analysis technology. This technology includes a local region segmentation signal processing quantification algorithm technology that reflects the response characteristics. That is, the machine learning-based quantitative analysis technology removes the noise 2D plane by constructing the optimal quantification curve based on machine learning, and automatically extracts parameters for sigmoidal quantitative analysis.

In addition, in an embodiment, an optimal response function model is constructed through an optimal machine learning-based regression analysis for multiple fluorescent substances in the fluorescence detection scanning system, which has superior dynamic range and coefficient of determination (R2) compared to existing technologies. , the limit of detection can be implemented.

A high-speed and high-sensitivity multi-channel fluorescence detection scanning system according to another aspect of the present invention for solving the above technical problem includes: a multi-fluorescence detection optical module having a light emission-photodetection scan structure; a two-dimensional driving stage supporting the plane movement of the multi-fluorescence detection optical module; and a system control board for controlling operations of the multiple fluorescence detection optical module and the two-dimensional driving stage.

In one embodiment, the multi-fluorescence detection optical module includes a light source, an objective lens disposed on an irradiation light path of the light source, a mirror disposed on the light path of the light source, and a well reflecting light passing through the mirror ( well) type sample platform, a detector for receiving light reflected from the sample platform through reflection of the mirror, and a condensing lens disposed between the mirror and the detector.

In one embodiment, the multi-fluorescence detection optical module includes a light source, an objective lens disposed on an irradiation light path of the light source, a well-type sample platform that reflects the light of the light source, and is disposed next to the light source and a detector for detecting light reflected from the sample platform, a condensing lens disposed between the sample platform and the detector, and an emission filter disposed between the condensing lens and the detector.

In an embodiment, the light source has a structure in which light emitting diodes (LEDs) accommodating multiple fluorescence wavelength bands for each channel are arranged in an array form.

In an embodiment, the two-dimensional driving stage may include: a first slide bar extending in a first direction or a transverse direction; a first slide bar joint coupled to the first slide bar and moving along the first slide bar; a first driving unit coupled to one end of the first slide bar and installed to move the first slide bar joint; a second slide bar having one end coupled to the first slide bar joint and extending in a second direction orthogonal to the first direction or a longitudinal direction; a second slide bar joint coupled to the second slide bar and moving along the second slide bar; and a second driving unit coupled to the other end of the second slide bar and installed to move the second slide bar joint, wherein the first slide bar includes a plurality of first bars. spaced apart from each other in a first direction and a third direction perpendicular to the second direction, the first slide bar joint is slidably coupled to the plurality of first bars, and the second slide bar includes the plurality of second bars Second bars are spaced apart in the third direction, the second slide bar joint is slidably coupled to the plurality of second bars, and one surface of the second slide bar joint in the third direction has The multiple fluorescence detection optical modules are combined.

In an embodiment, the system control board controls the operation of the plurality of light sources of the multi-fluorescence detection optical module, controls the operation of the two-dimensional driving stage through operation control of a step motor, and the multi-fluorescence detection optical module The detection signal of the plurality of detectors of the module is received, and the received detection signal is transmitted to the diagnostic system.

In an embodiment, the main control unit of the system control board repeats scanning a plurality of times for a plurality of microregions in which array wells containing a liquid sample are horizontally and vertically divided.

In an embodiment, the main controller detects a temperature change using a sensor, and performs temperature compensation on the detection signals of the plurality of detectors based on a preset reference curve according to the temperature change.

In one embodiment, the main controller adjusts the focal position according to the effective range of fluorescence dispersion according to the detection reaction degree of the liquid sample distributed in the three-dimensional space.

In an embodiment, the diagnosis system determines an effective interval for a nonlinear or curved surface of 2D or 3D measurement data including the detection signal through a regression analysis-based machine learning algorithm or a quantitative analysis algorithm.

Using the above-described fluorescence detection scanning system, it is possible to provide a 6-channel quantitative analysis fluorescence detection scanning system for liquid samples applicable to advanced qPCR, real-time PCR, or digital PCR fields. In other words, high speed and high light efficiency can be increased through optical design, and the flexible design structure of the optical system and drive system has the ability to adapt to diagnostic system equipment and has a high signal-to-noise ratio (SNR). The detection signal of the output can be obtained, and through this, the detection limit and sensitivity can be secured as a fluorescence detection scanning system mounted on an integrated diagnostic PCR to ensure on-site performance and speed.

In addition, according to the present invention, when a predetermined target area is divided into several divided areas, it is possible to increase the detection density of the fluorescent material per small unit area. That is, in the case of the existing technology, if there is one mutagenic substance out of 100 target areas, false negatives can be caused at a ratio of 1/100, whereas the micro-region detection method in which the target area is divided into 100 is measured. In the present embodiment, the mutable substance can be detected with a probability of 1/1, so that the detection probability can be dramatically increased, thereby improving the detection efficiency and precision of the system.

In addition, according to the present invention, the signal-to-noise ratio (SNR) of the total detection signal is dramatically improved by collecting a lot of data to remove noise included in the detection signal and implementing an optimal noise removal algorithm through machine learning. can be increased to That is, according to the present invention, it is possible to minimize the error of quantitative calculation by applying a machine learning-based quantitative analysis technique to find a nonlinear model function suitable for efficient noise removal and waveform analysis of the detected measurement signal, and thereby The reproducibility and accuracy of the quantification of the system can be improved.

1 is a perspective view of a fluorescence detection scanning system according to an embodiment of the present invention.
FIG. 2 is a right side view of the fluorescence detection scanning system of FIG. 1 .
FIG. 3 is a rear view of the fluorescence detection scanning system of FIG. 1 .
4 is a perspective view illustrating a state in which an upper cover is removed from the fluorescence detection scanning system of FIG. 1 .
5 is a schematic diagram illustrating a state in which the upper cover and the side body case are removed from the fluorescence detection scanning system of FIG. 1 .
FIG. 6 is a schematic view of the fluorescence detection scanning system of FIG. 5 as viewed from the lower side.
FIG. 7 is an exemplary diagram of a two-dimensional driving stage that can be employed in the fluorescence detection scanning system of FIG. 1 .
FIG. 8 is an exemplary diagram of a configuration for high-sensitivity confocal optical detection that can be employed in the fluorescence detection scanning system of FIG. 7 .
FIG. 9 is an exemplary diagram of another two-dimensional driving stage that can be employed in the fluorescence detection scanning system of FIG. 1 .
FIG. 10 is an exemplary diagram of another configuration for high-sensitivity confocal optical detection that can be employed in the fluorescence detection scanning system of FIG. 7 .
11 is a view for explaining a process of selecting an excitation light source and setting a photodetector that can be employed in the fluorescence detection scanning system of FIG. 1 .
12 is an exemplary diagram of a configuration and operating system of a control board that can be employed in the fluorescence detection scanning system of FIG. 1 .
13 is an exemplary diagram of a configuration method of a scanning scheduler state machine that can be employed in the fluorescence detection scanning system of FIG. 1 .
FIG. 14 is a conceptual diagram for explaining a micro-region division scan and signal processing process that can be employed in the fluorescence detection scanning system of FIG. 1 .
FIG. 15 is an exemplary view for explaining a process of improving detection density by whole-area detection and fine-region segmentation detection that can be employed in the fluorescence detection scanning system of FIG. 1 .
16 and 17 are exemplary diagrams for explaining a machine learning-based quantification analysis process that can be employed in the fluorescence detection scanning system of FIG. 1 .
18 is an exemplary view for explaining a rapid diagnosis kit that can be employed in the fluorescence detection scanning system of FIG. 1 .
19 is an exemplary diagram of the internal multi-biomarker array structure of the rapid diagnostic kit of FIG. 18 .
20 is a graph illustrating a correlation between a biomarker concentration detected in the biomarker array of FIG. 19 and a photodetection value.
FIG. 21 is an exemplary diagram of an application process of variable output irradiation light when scanning the biomarker array of FIG. 19 .
22 is an exemplary diagram illustrating characteristic curves of variable output irradiated light with respect to the biomarker array of FIG. 19 .
FIG. 23 is a diagram for explaining a detection process for a biomarker of a wide concentration using the characteristic curves of FIG. 22 .
24 is a block diagram of a control system employable in the fluorescence detection scanning system of FIG. 1 .

Since the present invention can have various changes and can have various embodiments, specific embodiments are illustrated in the drawings and described in detail in the detailed description. However, this is not intended to limit the present invention to specific embodiments, and it should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention.

Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a perspective view of a fluorescence detection scanning system according to an embodiment of the present invention. FIG. 2 is a right side view of the fluorescence detection scanning system of FIG. 1 . FIG. 3 is a rear view of the fluorescence detection scanning system of FIG. 1 . 4 is a perspective view illustrating a state in which an upper cover is removed from the fluorescence detection scanning system of FIG. 1 . FIG. 5 is a schematic diagram illustrating a state in which an upper cover and a side body case are removed from the fluorescence detection scanning system of FIG. 1 . And, FIG. 6 is a schematic view of the fluorescence detection scanning system of FIG. 5 as viewed from the lower side.

1 to 6 , a high-speed and high-sensitivity fluorescence detection scanning system (hereinafter, simply referred to as a fluorescence detection scanning system) 100 according to the present embodiment has a width of 20 cm to 40 cm, a length of 20 cm to 40 cm, A rectangular-shaped housing having a height of about 20 cm to 40 cm is provided. The housing may include an upper cover, a side body case, and a lower support that are fastened to each other to have a rectangular parallelepiped shape.

In addition, the fluorescence detection scanning system 100 includes a multiple fluorescence detection optical module 10 , a two-dimensional driving stage 20 and a system control board 30 therein.

The fluorescence detection scanning system 100 according to the present embodiment is a system mounted on an integrated diagnostic PCR (Polymerase Chain Reaction) to ensure on-site performance and speed, and its detection limit and sensitivity can be secured.

A specific configuration for implementing a fluorescence detection scanning system that improves the detection limit and sensitivity to ensure on-site performance and rapidity will be mentioned in the detailed description below.

FIG. 7 is an exemplary diagram of a two-dimensional driving stage that can be employed in the fluorescence detection scanning system of FIG. 1 . FIG. 8 is an exemplary diagram of a configuration for high-sensitivity confocal optical detection that can be employed in the fluorescence detection scanning system of FIG. 7 .

7 (a) and (b) and 8, the fluorescence detection scanning system according to the present embodiment includes a multiple fluorescence detection optical module (hereinafter also simply referred to as an optical module) 10 and the optical module 10. A two-dimensional driving stage 20 that supports the in-plane movement of

The optical module 10 is designed and manufactured to have various optical paths to enable high speed. The optical module 10 includes an LED, an objective lens, a detector, an emission filter, a condensing lens, a well-type sample platform 2, and a mirror. It is possible to operate so that the beam irradiated from the LED through the objective lens passes through the mirror, is reflected by the well, is reflected back from the mirror, and then is detected by the detector through the condensing lens and the irradiation filter.

In other words, the optical module 10 includes a light source such as a light emitting diode (LED), an objective lens disposed on the irradiation light path of the light source, a mirror disposed on the light path of the light source, and a well reflecting light passing through the mirror. (well) type sample platform 2, a detector that receives light reflected from each well of the sample platform 2 through reflection of a mirror, and a condensing lens disposed between the mirror and the detector do.

That is, the optical module 10 may be implemented in an upward emission-side detection scan method on the sample platform 2 in the form of an array for high-sensitivity multi-detection. That is, it is possible to detect fluorescence in PCR in such a way that light is emitted from top to bottom, and light reflected from bottom to top and reflected back from the mirror is detected from the side of the system device.

In this embodiment, the optical module 10 may have an arrangement structure of a fluorescence detection optical system for simultaneous measurement of six tubes. The light source may have a structure in which light emitting diodes (LEDs) accommodating multiple fluorescence wavelength bands for each channel are arranged in an array form. The detector may include, but is not limited to, a detector photodiode including a photodiode, an avalanche photodiode (APD), a silicon photomultiplier (SiPM), and a multi-pixel photon. A multi-pixel photon counter, a photo multiplier tube (PMT), or the like may be used.

The two-dimensional driving stage 20 includes a first slide bar extending in a first direction or a transverse direction; a first slide bar joint coupled to the first slide bar and moving along the first slide bar; a first driving unit coupled to one end of the first slide bar and installed to move the first slide bar joint; a second slide bar having one end coupled to the first slide bar joint and extending in a second direction (or longitudinal direction) perpendicular to the first direction; a second slide bar joint coupled to the second slide bar and moving along the second slide bar; and a second driving unit coupled to the other end of the second slide bar and installed to move the second slide bar joint.

Here, in the first slide bar, a plurality of first bars are spaced apart from each other in a third direction orthogonal to the first direction and the second direction, respectively, and the first slide bar joint slides on the plurality of first bars. possible to combine

In addition, in the second slide bar, a plurality of second bars are disposed to be spaced apart in a third direction, and the second slide bar joint is slidably coupled to the plurality of second bars.

In the two-dimensional driving stage 20 as described above, the multiple fluorescence detection optical module 10 may be coupled to one side or one side of the second slide bar joint in the third direction. The third direction corresponds to a direction from the two-dimensional driving stage 20 toward a well-type sample platform (hereinafter also referred to as a 'well plate' for short) 2 .

According to the configuration of the optical module 10 of the present embodiment, it is possible to secure a structure that easily responds to a specification change of the well plate 2 containing the liquid sample.

FIG. 9 is an exemplary diagram of another two-dimensional driving stage that can be employed in the fluorescence detection scanning system of FIG. 1 . FIG. 10 is an exemplary diagram of another configuration for high-sensitivity confocal optical detection that can be employed in the fluorescence detection scanning system of FIG. 9 . 11 is an exemplary view for explaining a process of selecting an excitation light source and setting a photodetector that can be employed in the fluorescence detection scanning system of FIG. 1 .

9 (a) and (b) and 10, the fluorescence detection scanning system according to the present embodiment is a multi-fluorescence detection optical module 10 and a two-dimensional support for the plane movement of the optical module 10. A driving stage 20 is provided. The multiple fluorescence detection optical module 10 may be implemented in an upward emission-phase detection scan method for high-sensitivity detection. That is, the fluorescence of PCR can be detected by emitting light from top to bottom and detecting the light reflected from bottom to top from the top side.

The optical module 10 is designed and manufactured to have various optical paths capable of high speed. The optical module 10 includes a light source including an LED, an objective lens, a detector photodiode, an emission filter, a condensing lens, and a well-type sample platform ( 2), and when the beam irradiated from the LED through the objective lens is reflected from the well, it can be configured to detect a signal with a detector photodiode (detector PD) installed next to the LED.

Here, light reflected from the well may be incident on the detector photodiode facing the sample platform 2 through a condensing lens and an emission filter.

The light source may have a structure in which light emitting diodes (LEDs) accommodating multiple fluorescence wavelength bands for each channel are arranged in an array as shown in FIG. 11A . In order to form multiple fluorescence wavelength bands for each channel, the LED light source includes the first channel FAM (450-490 nm), the second channel HEX (515-535 nm), the third channel Texas Red (560-590 nm), and the fourth channel. An excitation dye for Cy5 (620 to 650 nm), a fifth channel Quasar 705 (673 to 684 nm), and a sixth channel blank light may be used.

Similarly, as shown in FIG. 11B , the detector may have a structure in which photodiodes for detecting signals of multiple fluorescence wavelength bands for each channel are arranged in an array form. In order to detect signals in multiple fluorescence wavelength bands for each channel, the detector includes the first channel FAM (510-530 nm), the second channel HEX (560-580 nm), the third channel Texas Red (610-650 nm), and the fourth channel. An emission dye or a detection dye for the channel Cy5 (675 to 690 nm), the fifth channel Quasar 705 (705 to 730 nm), and the sixth channel blank light may be used.

In this embodiment, the optical module 10 may have an arrangement structure of a fluorescence detection optical system for simultaneous measurement of six tubes.

Since the two-dimensional driving stage 20 is substantially the same as the corresponding components with reference to FIGS. 7 and 8 , a detailed description thereof will be omitted.

According to the optical module of this embodiment, it is possible to secure a structure that easily responds to variations in the specifications of the well plate containing the liquid sample.

12 is an exemplary diagram of a configuration and operating system of a control board that can be employed in the fluorescence detection scanning system of FIG. 1 .

Referring to FIG. 12 , the system control board (hereinafter also referred to as a 'control board') 30 according to the present embodiment includes a main control unit (MCU, 50), an input unit 62, and a drive system control unit 64. , a power supply unit 66 and an output unit 68 are provided. The input unit 62 corresponds to a detector including a plurality of photodiodes (PD1 to PD6, etc.). The drive system control unit 64 includes a step motor or an actuator that performs a function corresponding thereto. The power supply unit 66 includes a main power such as a commercial power supply or a secondary battery. The output unit 68 may include an array of a plurality of light emitting diodes (LED1 to LED6, etc.).

The control board 30 controls the operation of the plurality of light sources of the multi-fluorescence detection optical module, controls the operation of the two-dimensional driving stage through operation control of the step motor, and the plurality of detectors of the multi-fluorescence detection optical module Receives the detection signal of the two and transmits the received detection signal to the diagnostic system (70). The diagnosis system 70 may determine an effective section for a nonlinear or curved surface of 2D or 3D measurement data including a detection signal through a regression analysis-based machine learning algorithm or a quantitative analysis algorithm.

In addition, the main control unit 50 of the control board 30 operates to repeat the scanning a plurality of times for the array wells containing the liquid sample, that is, a plurality of microregions in which the well plate is divided horizontally and vertically. can

In addition, the main controller 50 detects a temperature change around the well plate using a sensor, and performs temperature compensation on the detection signals of the plurality of detectors based on a preset reference curve according to the detected temperature change. .

In addition, the main controller 50 may adjust the focal position according to the effective range of fluorescence dispersion according to the detection reaction degree of the liquid sample distributed in the three-dimensional space. In that case, by precisely adjusting the focal position in combination with the precise scan control operation of the two-dimensional driving stage, the detection limit and sensitivity of the fluorescence detection scanning system can be improved, and the detection efficiency and precision can be improved.

13 is an exemplary diagram of a configuration method of a scanning scheduler state machine that can be employed in the fluorescence detection scanning system of FIG. 1 .

The fluorescence detection scanning system according to the present embodiment may use a scheduler for precise scan control. The scheduler may be mounted on the main control unit of the system in the form of firmware (F/W).

The scheduler divides the array well into 8 sub-regions or micro-regions for scanning 96 wells (8 rows, 12 columns) containing liquid samples and applies them to perform at least 4 round-trip scans per column. can be configured.

As shown in FIG. 13, the scheduler moves the position of the optical module by the two-dimensional driving stage (access_XY) (S81), the on/off operation of the light source with respect to the third column from the 0th column ( After performing a series of processes for S82) and the detector's photodetection operation (S83) (S81 to S84), move to the fourth column, and for the fourth column to the seventh column, optical by a two-dimensional driving stage After performing a series of processes for the position movement of the module (access_XY) (S85), the on/off operation of the light source (S86), and the photodetection operation of the detector (S87) (S85 to S88), Move to the 8th column and move the position of the optical module by the two-dimensional driving stage with respect to the 8th column to the 11th column (access_XY) (S89), the on/off operation of the light source (S90) And after performing a series of processes for the photodetection operation (S91) of the detector (S89 to S92), move to the twelfth column and move the position of the optical module by the two-dimensional driving stage with respect to the twelfth column (access_XY) (S93), after performing a series of processes for the on/off operation of the light source (S94, S95) and the light detection operation of the detector that are repeated up to a preset row (12 rows) (S93 to S95) ), may be set to move to the initial position (S96).

When a detailed scan is performed according to the above scheduler, the time required for one round trip is about 20 seconds, and the scanning time for one column is about 80 seconds according to the calculation process of 20 seconds/round trip * 4 round trips/column, The total scanning time can be about 960 seconds (16 minutes) according to the calculation process of 80 seconds/row * 12 columns.

FIG. 14 is a conceptual diagram for explaining a micro-region division scan and signal processing process that can be employed in the fluorescence detection scanning system of FIG. 1 . FIG. 15 is an exemplary view for explaining a process of improving detection density by whole area detection and fine area segmentation detection that can be employed in the fluorescence detection scanning system of FIG. 1 .

14 and 15 , the fluorescence detection scanning system according to the present embodiment performs fluorescence detection scanning on the sample from the liquid sample contained in the well of the well plate 2 by adjusting the focal position and dividing the fine region.

That is, in the present embodiment, the fluorescent signal is detected by scanning the measurement target region in a fine region division method in which the measurement target region is subdivided into fine units, rather than detecting the fluorescent material in a predetermined volume of the liquid sample by one measurement. When a certain target area is divided into several divided areas, it is possible to increase the detection density of the fluorescent material per small unit area. As exemplified by the comparative example of FIG. 15(a), when there is one mutagenic substance in the target area 100, it may cause false negatives at a ratio of 1/100 in the corresponding image region 40a of the measurement data. On the other hand, as exemplified by the embodiment of FIG. 15( b ), the fine-region division detection technique for measuring the target area by dividing it into 100 parts is detected with a probability of 1/1 in the corresponding image region 40 of the measurement data. This can dramatically increase the detection probability. As described above, according to the scanning method of the fine region division method according to the present embodiment, detection efficiency and precision can be improved.

16 and 17 are exemplary views for explaining a machine learning-based quantification analysis process that can be employed in the fluorescence detection scanning system of FIG. 1 .

The fluorescence detection scanning system according to the present embodiment may include a process of arranging and analyzing the scan line signal data of the rapid diagnostic kit. For example, as shown in FIG. 16 , a graph connecting scan line signal data values is first converted into an approximate curve using the average of continuous signal values of a predetermined section. This approximate curve conversion process may be performed by setting 3 to 12 data points of continuous data as section data and calculating the average of the section data (see FIGS. 16 (a) and (b)). According to the approximate curve conversion process, the curve of the scan line signal data may be smoothed and the signal value may be slightly lowered.

Next, a region of effect (ROE) of scan line signal data having a predetermined width or constant intensity is determined based on a peak value in each section aligned with a peak value of a region of interest (ROI). ROI1 may be the region of interest of the control signal line and ROI2 may be the region of interest of the test signal line, and vice versa.

In the above process, the microarray of the rapid diagnostic kit is generated at the boundary of each signal line when the spot area of the laser beam moves along the scan line, that is, when it approaches or passes over the data signal line or the control signal line. This is to remove virtually invalid data, and through this process, reliable quantification data can be provided by limiting the data to be analyzed to only higher-quality data and reducing the capacity.

Next, the average (ROE1, ROE2) of the data values for each valid area for each scan line may be obtained, and final data for quantification may be output (refer to FIG. 16(c) ). The final data is not limited to averaging the data for each valid region, and various methods of taking a median value or averaging values in the median range are alternatively available.

As described above, the fluorescence detection scanning system according to the present embodiment can perform analysis by applying a local region division signal processing quantification algorithm that reflects the reaction characteristics after scanning a sample on the microarray of the rapid diagnostic kit in the form of a strip bar. . This analysis process removes noise through machine learning on the collected data to remove noise included in the detection signal, that is, by applying the optimal noise removal algorithm based on machine learning, the signal-to-noise ratio of the total detection signal (signal-to-noise ratio) noise ratio: SNR) can be significantly increased. That is, the reliability of quantitative analysis can be improved by constructing an optimal quantification curve for a detection signal based on machine learning.

In addition, the fluorescence detection scanning system of this embodiment has excellent dynamic range, coefficient of determination (R2), and limit of measurement by constructing an optimal response function model through machine learning-based regression analysis for multiple fluorescent substances. detection) can be implemented. In addition, quantification analysis technology based on machine learning is applied to find a nonlinear model function suitable for efficient noise removal and waveform analysis of the detected measurement signal, thereby minimizing the error in quantitative calculation, and the reproducibility of quantification and accuracy can be improved.

For example, as shown in FIG. 17 , the machine learning-based quantification analysis process removes the noise 2D plane to construct an optimal quantification curve based on machine learning, and automatically adjusts parameters for sigmoidal quantification analysis. can be introduced to extract.

FIG. 18 is an exemplary view for explaining a rapid diagnosis kit that can be employed in the fluorescence detection scanning system of FIG. 1 . 19 is an exemplary diagram of the internal multi-biomarker array structure of the rapid diagnostic kit of FIG. 18 .

The fluorescence detection scanning system according to the present embodiment may be implemented to mount a rapid diagnosis kit. The rapid diagnosis kit 100 is, as shown in FIG. 18, a sample pad from one side to the other in the longitudinal direction on a plastic plate supporting them so that the overall shape of the pad or membrane is maintained inside the housing. ), a connection pad (conjugate pas), a membrane (membrane, 110), and an absorbent pad (absorbent pad) may be provided in a form installed to partially overlap in sequence. Here, a sample port is disposed on the sample pad.

As shown in FIGS. 18 and 19 , the liquid sample is introduced into the sample pad through a hole-shaped sample port formed in the housing, and then flows laterally from the sample pad to the absorption pad along the first direction D1 of the first arrow. The first direction D1 may correspond to the longitudinal direction of the membrane.

On the membrane 110 , a test line 220 and a control line 230 are formed in the order described in the direction from the sample pad to the absorption pad. In addition, one reflective reference line may be installed on the membrane 210 before or after the test line 220 and the control line 230 .

In this case, the reflective reference line may be installed in a form in which a thin metal film is formed on the surface of the membrane 210 to have a narrow width. The reason that the material is made of a metal film is to stably reflect the illumination light at a high rate by its metallic luster so that the scanning system can clearly detect the reflected light.

In addition, the thin film thickness is such that there is almost no level difference between the surface of the membrane 210 on which the metal film is installed and the surface of the metal film, so that when the focus is on the reflective reference line surface of the metal film, the surface of the membrane 210 on which the test line 220 is installed This was done to have the effect of focusing on

In addition, the narrow width of the metal film is such that the position in the lateral flow direction of the metal film is precisely defined close to one point within the range where the metal film is clearly detected by the scanning system and its analysis unit, and the metal film width increases and the This is to prevent this because if the number increases, there may be restrictions on the area of the test line, etc.

One reflective reference line may be installed at a predetermined distance apart from the front or rear of the test line, or one at the front and back. If it is, it is assumed that the test line 220 exists after or before the predetermined distance, the scanning system is moved by a certain distance with respect to the rapid diagnostic kit, and the optical signal obtained while the test line 220 is scanned secondarily can be used for analysis. have.

In addition, when the reflective reference line is installed one at the front and rear of the test line 220 or the control line 230, if the distance from the first reference line is moved by half the distance between the two reference lines confirmed in the first scan, the test line 220 position , and analysis can be performed by viewing the optical signal obtained during the secondary scan at this position as the optical signal of the test line.

The test line 220 has an array structure in which a plurality of capture material regions exist in a matrix form within the test line region. The array structure may be referred to as a microarray 221 . In that case, in order to obtain an optical signal while passing through each area constituting the matrix, the scan is performed in a second scan direction scan2 orthogonal to the first scan direction while moving at regular intervals in the first scan direction Scan1. In this case, it can be performed as if each die area on the wafer is scanned in a zigzag or inspected while passing through in a step and repeat method.

The control board of the scanning system may control the focusing of the irradiated light by controlling the optical system 240 disposed to face the detection surface direction. It may be implemented to collect signals over the entire test line 220 , average them, or select the strongest signal value for analysis in the analysis unit.

Meanwhile, in order to simplify the configuration and operation of the optical device of the scanning system, it is desirable to make the conditions such as irradiation light used for the first scan and the second scan the same, but in this embodiment, the characteristics of the reflective baseline and the immune response in the test line Since the characteristics of the resulting result may be optically different, in order to detect them under optimal conditions, the light signal is acquired by passing through the test line by the irradiation light used to detect the reflective baseline by the first scan and the second scan. It is possible to have different characteristics such as frequency of the irradiated light to be used when doing this, and it is also possible to consider using a different physical filter for detecting an optical signal or a filter technique by an electronic processing method. Similarly, in order to make the detection accuracy of the first scan and the second scan different, the scan speed, the sensitivity setting of the detector, amplification, etc. may be different.

The separation distance between the scan lines within the test line is a distance where there is no adverse effect due to mutual interference in the signal graph finally obtained by the analysis unit while scanning and the separation in terms of mechanical movement and optical aspect can be reliably made. It is desirable to shorten the test line to about 100 μm, and if a large separation distance of 1 mm or more occurs, the separation distance of the test line increases in proportion to the expansion and contraction of the membrane. There is a possibility of setting the wrong location out of the way, so it can be set taking this into consideration.

Although the above description is based on the test line, the same method may be applied to the control line 230 . However, if the detection of the result of the control line 230 can be obtained more clearly than the detection of the result of the test line, it will not be necessary to further provide a reflective reference line around the control line.

20 is a graph illustrating a correlation between a biomarker concentration detected in the biomarker array of FIG. 19 and a photodetection value. FIG. 21 is an exemplary diagram of an application process of variable output irradiation light when scanning the biomarker array of FIG. 19 . 22 is an exemplary diagram illustrating characteristic curves of variable output irradiated light with respect to the biomarker array of FIG. 19 . And FIG. 23 is a diagram for explaining a detection process for a biomarker of a wide concentration using the characteristic curves of FIG. 22 .

The profile of the detection signal at the biomarker spot obtained through the biomarker array-type rapid diagnostic kit according to this embodiment has a predetermined concentration in the test line (Test line, T) and the control line (Control line, C). according to the corresponding signal value. Accordingly, when the detection signal characteristics for each concentration of a specific biomarker are shown, the corresponding signal value in the test line increases as the concentration increases.

That is, looking at the signal variation according to the density range, when the concentration (Bd) is large, the measured signal change (Oi) is large, and when the concentration (Bd) is small, the measured signal change (Oi) is small When this is shown in a graph, it is as shown in FIG. 20 .

Referring to FIG. 20 , when the biomarker density is large, a high signal value can be obtained by increasing the intensity of the irradiation light (detection light intensity, optical intensity), and when the biomarker concentration is relatively low, the intensity of the irradiation light It can be seen that even if the (detection light intensity) is increased, a low signal value is obtained. Here, in the detection characteristic curve CG, the first level L1 represents a saturation level, the second level L2 represents a limit of detection, and DR represents an effective range.

As such, the correlation between the biomarker concentration and the photodetection value has an inverse S-curve shape. The detection characteristic curve of the inverse S-curve may correspond to a detection characteristic for each concentration in a predetermined detection light intensity and a predetermined amplification state.

Also, as shown in FIG. 21 , the scanning system according to the present embodiment can move in a zigzag form while changing the output of irradiation light when scanning the microarray of the rapid diagnosis kit.

That is, the control board of the scanning system controls the output (power, W) of the irradiation light in the width direction (scan1) of the microarray 221 in the rapid diagnosis kit or the rapid diagnosis kit, It is possible to control the output (power, W') of the irradiation light in the flow direction (scan2).

The first output W of the irradiated light in the width direction scan1 or the first direction may be controlled or output so that the output is strongest in the middle portion thereof. In addition, in the flow direction scan2 or the second direction orthogonal to the width direction, the second output W' of the irradiated light may be controlled or output to gradually increase in most areas. For example, when scanning starts from the first line r1 in the flow direction, the intensity of the irradiation light of the second line r2 adjacent to the first line r1 is the intensity of the irradiation light of the first line r1, the average intensity, It may be set or controlled to be greater than the line scan start intensity or the line scan last intensity.

Similarly, the intensity of the irradiation light of the third line r3 adjacent to the second line r2 in the flow direction scan2 is set to be greater than the intensity, average intensity, start intensity, or last intensity of the irradiation light of the second line r2, or can be controlled. In addition, the intensity of the irradiation light of the fourth line r4 adjacent to the third line r3 in the flow direction scan2 is greater than the intensity of the irradiation light of the third line r3, the average intensity, the line scan start intensity, or the line scan last intensity. can be set or controlled.

On the other hand, the intensity of the irradiation light of the fifth line r5 adjacent to the fourth line r4 in the flow direction scan2 is that of the irradiation light of the fourth line r4 when the fifth line r5 is the last line of scanning. The intensity, average intensity, line scan start intensity, or line scan end intensity may be set or controlled to be smaller than the intensity. In the above case, the intensity of the irradiated light of the fifth line r5 may be the same as the intensity of the corresponding irradiated light of the first line r1 , the second line r2 , or the third line r3 .

When the intensity of the variable output irradiation light of the above-described optical scanning is schematically represented as a graph, the power before transmission (excitation power) and power after transmission (emission power) are the same as shown in FIG. 22 . The power before transmission represents the light intensity before the irradiation light of the light emitting device of the scanning system does not pass through the microarray or the sample in the microarray, and the power after transmission represents the intensity of the detection light after the irradiation light passes through the microarray or the sample . As shown in FIG. 8 , the pre-transmission power may be set or controlled to be gradually increased during most of the scanning process in the form of a waveform.

In addition, as shown in FIGS. 22 and 23, the fluorescence detection scanning system according to the present embodiment uses variable excitation light power, that is, detection characteristic curves S1, S2, and S3 by variable output irradiated light, to achieve a wide range of concentrations. It is possible to generate a new detection characteristic curve (S10) with an extended dynamic range for the detection of multiple biomarkers of

The detection characteristic curve S10 with the extended dynamic range can be obtained by correcting the detection characteristic curves of the variable output irradiation light for each multi-biomarker concentration range obtained by scanning the microarray using the variable output irradiation light.

For example, in this embodiment, an extended detection characteristic curve in the form of a single curve, that is, an extended detection characteristic curve having a plurality of inflection points, can be obtained by removing the deviation of the detection characteristic curves for each concentration range, and this detection characteristic curve (S10) may have a dramatically extended dynamic range (dr1+dr2+dr3) in comparison with the dynamic range before expansion (dr1) of the detection characteristic curve for each concentration in a constant excitation light power and a constant amplification state. Therefore, according to this embodiment, it is possible to effectively provide a high-speed and high-sensitivity multi-channel fluorescence detection scanning system.

24 is a block diagram of a control system employable in the fluorescence detection scanning system of FIG. 1 .

Referring to FIG. 24 , the control board 290 of the fluorescence detection scanning system according to the present embodiment is a type of the system control board of FIG. 5 , and includes an input unit 291 and an output unit 292 . , a main control unit (MCU, 293), a laser diode power supply (LD Power, 294), an amplifier (295), an analog-to-digital converter (ADC, 296), and a motor driving unit (Moter driver, 297), It may be connected to a laser diode (LD), a photodiode (PD), a first motor and a second motor of the scanning system. The first motor may be provided for lateral movement and the second motor may be provided for longitudinal movement.

The control board 290 may be connected to the display device 298 , and the display device 298 may include a touch screen for both input and output, such as a touch liquid crystal display (Touch LCD).

The main control unit 293 may operate to extract the photodiode data at regular intervals by decomposing the unit reference signal into 4096 steps at 12-bit resolution, for example. In addition, the main control unit 293 sets the frequency for the digital signal through pulse width modulation (PWM) control, and the pulse width or duty cycle adjusts the amplitude of the signal to adjust the intensity of the laser diode. have. According to this configuration, it is possible to significantly improve the conversion speed in the analog-to-digital conversion system compared to the control unit with 4-bit or 8-bit resolution.

The main control unit 293 may be referred to as a control unit, a control unit, a central processing unit, or the like, and may be implemented as a microprocessor, a microcomputer, or the like.

Also, the main control unit 293 may store an analysis program for noise reduction and effective information acquisition in a memory, and may execute the analysis program through the main control unit 293 connected to the memory.

The analysis program may include a scanning schedule algorithm, a photodiode (PD) detection voltage accumulation and analysis algorithm, and the like.

A scanning schedule algorithm is driven in the y-axis direction to form an autofocus arrangement in a first step, wherein the y-axis is decomposed into sections larger than 100 to 200 sections; a second step of returning to the y-axis origin; a third step of driving to form an autofocus arrangement in the x-axis direction, wherein the x-axis is decomposed into sections larger than 40 to 80 sections; a fourth step of returning to the origin of the x-axis; a fifth step of moving in the y-axis direction to start scanning the test line and moving the micro-section in the y-axis direction, wherein, when moving the micro section, the autofocus arrangement formed in the first step is used-; And it may include a series of processes of the sixth step of continuously applying the fifth step to complete the scan from the x-axis origin to the end point and return to the x-axis origin.

And, the PD detection voltage accumulation and analysis algorithm is a two-dimensional memory array A[n][ m]; The second step of storing the voltage (V) of the signal detected by the photodiode or the light receiving element in a specific memory area of the two-dimensional memory array A[n][m] - Here, the base voltage of the opaque section corresponds to 0 or a predetermined voltage -; A voltage one-dimensional memory array (B[n]) for storing the sum of voltages from the first (x[0]) to the last (x[m-1]) of the x-axis with respect to the i-th (y[i]) of the y-axis a third step of generating ]); a fourth step of summing the voltages for x[0] to x[m-1] with respect to the y[i] and storing them in a specific one-dimensional memory array (B[j]); A specific one-dimensional memory array (B[j]) by summing the voltages for x[0] to x[m-1] with respect to [0] to [n-1] of the y-axis as a step of repeating the fourth step a fifth step of storing each in ; a sixth step of calculating a detection sensitivity (Ii) by dividing each B[j] by the corresponding area (Si); and a seventh step of selecting the maximum value (Imax) of the detection sensitivity as the effective determination reference value.

Here, the sum of the detection voltages in the transparent section of the aforementioned microarray can be calculated by Equation 1 below.

In addition, each section detection sensitivity can be calculated by the following Equation (2).

According to this embodiment, the photodiode output voltage, which is a signal on the microarray, is obtained in an appropriate range according to the size of the irradiation light set for each biomarker concentration. can provide Here, the control board 290 may have a form coupled to a base plate or a support frame of the fluorescence detection scanning system according to implementation.

In addition, the fluorescence detection scanning system using the rapid diagnostic kit of this embodiment is associated with diseases in which various antigens and biomarkers exist, such as specific diseases, such as sexually transmitted diseases and cardiovascular diseases, and has various measurement methods and different concentration ranges. A lateral flow-based rapid diagnostic kit that can simultaneously detect dozens or more biomarkers in the form of a microarray on a single device is available, and a line-shaped signal region is integrated using microspotting technology It can be detected in the form of a fixed spot arrangement, and high-sensitivity fluorescence detection of 1 to 10 pg/ml level can be performed by applying the optimized quantitative algorithm technology for individual signals.

In addition, the fluorescence detection scanning system according to the present embodiment includes a variable power excitation mode and a variable gain detection mode ( By applying the variable gain detection mode technology, it is possible to effectively detect multiple biomarkers by extending the dynamic range.

Although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art can variously modify and change the present invention within the scope without departing from the spirit and scope of the present invention as set forth in the following claims. You will understand that there is

Claims (10)

a multiple fluorescence detection optical module having a light emission-photodetection scan structure;
a two-dimensional driving stage supporting the plane movement of the multi-fluorescence detection optical module; and
and a system control board for controlling operations of the multiple fluorescence detection optical module and the two-dimensional driving stage. The method according to claim 1,
The multi-fluorescence detection optical module includes a light source, an objective lens disposed on an irradiation light path of the light source, a mirror disposed on the light path of the light source, and a well-type sample that reflects light passing through the mirror A fluorescence detection scanning system comprising: a platform; a detector for receiving light reflected from the sample platform through reflection of the mirror; and a condensing lens disposed between the mirror and the detector. The method according to claim 1,
The multi-fluorescence detection optical module includes a light source, an objective lens disposed on an irradiation light path of the light source, a well-type sample platform that reflects the light of the light source, and is disposed next to the light source and is reflected from the sample platform A fluorescence detection scanning system comprising: a detector for detecting the emitted light; a condensing lens disposed between the sample platform and the detector; and an emission filter disposed between the condensing lens and the detector. 4. The method of claim 2 or 3,
The light source is a fluorescence detection scanning system having a structure in which light emitting diodes (LEDs) accommodating multiple fluorescence wavelength bands for each channel are arranged in an array form. The method according to claim 1,
The two-dimensional driving stage,
a first slide bar extending in a first direction or a transverse direction;
a first slide bar joint coupled to the first slide bar and moving along the first slide bar;
a first driving unit coupled to one end of the first slide bar and installed to move the first slide bar joint;
a second slide bar having one end coupled to the first slide bar joint and extending in a second direction orthogonal to the first direction or in a longitudinal direction;
a second slide bar joint coupled to the second slide bar and moving along the second slide bar; and
a second driving unit coupled to the other end of the second slide bar and installed to move the second slide bar joint; and
The first slide bar includes a plurality of first bars spaced apart from each other in a third direction orthogonal to the first direction and the second direction, and the first slide bar joint includes a plurality of first bars slidably coupled to the
The second slide bar includes a plurality of second bars spaced apart from each other in the third direction, and the second slide bar joint is slidably coupled to the plurality of second bars;
A fluorescence detection scanning system in which the multiple fluorescence detection optical module is coupled to one surface of the second slide bar joint in the third direction. The method according to claim 1,
The system control board controls the operation of the plurality of light sources of the multi-fluorescence detection optical module, controls the operation of the two-dimensional driving stage through operation control of the step motor, and the plurality of detectors of the multi-fluorescence detection optical module A fluorescence detection scanning system that receives a detection signal of the two and transmits the received detection signal to a diagnostic system. 7. The method of claim 6,
The main control unit of the system control board repeats scanning a plurality of times for a plurality of microregions in which array wells containing a liquid sample are divided horizontally and vertically. 8. The method of claim 7,
The main controller detects a temperature change using a sensor, and performs temperature compensation on the detection signals of the plurality of detectors based on a preset reference curve according to the temperature change. 8. The method of claim 7,
The main control unit adjusts the focal position according to the effective range of fluorescence dispersion according to the detection reaction degree of the liquid sample distributed in the three-dimensional space, the fluorescence detection scanning system. 7. The method of claim 6,
The diagnostic system determines an effective section for a nonlinear or curved surface of the 2D or 3D measurement data including the detection signal through a regression analysis-based machine learning algorithm or a quantitative analysis algorithm.

Images